vLLM x VastAI/Recipes
Qwen

Qwen/Qwen3-4B-Thinking-2507

Qwen3-4B-Thinking-2507 is a large language model with 4B parameters, capable of performing text-to-text generation with a wide range of tasks. The model is trained on a large amount of data, and is capable of handling long-form inputs.

Compact Qwen3-4B-Thinking-2507 model

View on HuggingFace
dense4B131,072 ctxvLLM 0.17.0text
Guide

Overview

Qwen3-4B-Thinking-2507 is a large language model with 4B parameters, capable of performing text-to-text generation with a wide range of tasks. The model is trained on a large amount of data, and is capable of handling long-form inputs.

Prerequisites

  • vLLM version: 0.17.0
  • Config: GPUs = 1x VA16, Precision = FP8, TP = 2, max-model-len = 64K

Example config only. Refer to above for others.

Start Docker Container

docker run \
  --privileged=true \
  --name vllm_service \
  --shm-size=256g \
  --ipc=host \
  -p 8000:8000 \
  -it \
  -v ~/.cache/huggingface:/root/.cache/huggingface \
  harbor.vastaitech.com/ai_deliver/vllm_vacc:latest \
  bash

Launching the Server

FP8 on 1x VA16

vllm serve Qwen/Qwen3-4B-Thinking-2507-FP8 \
  --trust-remote-code \
  --tensor-parallel-size 2 \
  --max-model-len 65536 \
  --enable-auto-tool-choice --tool-call-parser hermes \
  --enforce-eager

Client Usage

from openai import OpenAI

client = OpenAI(api_key="EMPTY", base_url="http://localhost:8000/v1")
resp = client.chat.completions.create(
    model="Qwen/Qwen3-4B-Thinking-2507-FP8",
    messages=[{"role": "user", "content": "Explain gated delta networks in one paragraph."}],
    max_tokens=512,
)
print(resp.choices[0].message.content)

References

Qwen/Qwen3-4B-Thinking-2507 | vLLM Recipes